CHEMOTACTIC POTENCY OF RECOMBINANT HUMAN NEUTROPHIL ATTRACTANT/ACTIVATION PROTEIN-l (INTERLEUKIN-8) FOR POLYMORPHONUCLEAR LEUKOCYTES OF DIFFERENT SPECIES Antal Rot In order to establish the species cross-reactivity of the human neutrophil attractant/ activation protein-l (interleukin-8, NAP-l/IL-g) and find which experimental species are responsive to the human cytokine, blood polymorphonuclear leukocytes (PNMLs) were isolated from chicken, dog, goat, guinea-pig, monkey, mouse, pig, rabbit, and rat and their in vitro migration in response to this cytokine was investigated. PMNLs from all of the tested species migrated in response to recombinant human NAP-l/IL-8 (rhNAP-l/IL-a). The potency of rhNAP-l/IL-8 for the PMNLs of different species varied and was considerably lower than its potency for human cells. The morphological study combined with the leukocyte enumeration in the intradermal rhNAP-l/IL-8 injection sites established an in vivo proinflammatory potency of rhNAP-l/IL-8 for rabbit and rat that was comparable to the observed in vitro chemotactic potency of rhNAP-l/IL-8 for neutrophils of these species. Copyright o 1991 by W.B. Saunders Company
The polymorphonuclear neutrophil granulocyte (PMNL or neutrophil) is the predominant human blood leukocyte. The appearance and accumulation of neutrophils in the tissues is one of the morphological hallmarks of acute inflammation. Upon stimulation the circulating neutrophils marginate, adhere to the endothelial lining of the postcapillary venules, and, driven by the gradient of the chemotaxin, migrate across the vessel wall into the tissues where at the site of inflammation they are triggered by high concentrations of inflammatory mediators to generate reactive oxygen radicals and secrete the contents of lysosomal granules.’ Whereas most of the known leukocyte chemotaxins attract both neutrophils and monocytes, the recently described neutrophil attractant/activation protein-l (interleukin-8, NAP-l/IL+ attracts neutrophils, lymphocytes, and basophils, but not monocytes.2-4 Thus, in contrast to the formylated peptides, C5a and LTB,, NAP-l/IL-8 could be the mediator of the inflammatory
Sandoz Forschungsinstitut, Copyright 0 1991 by W.B. 1043-4666/91/0301-0008$05.00/0 KEY WORDS: specificity
CYTOKINE,
Brunner Saunders
Str. 59, Vienna Company
chemotaxislinterleukin-8/IL-g/neutrophils/species
Vol. 3, No. 1 (January),
1991: pp 21-27
A-1235,
Austria.
responses dominated by neutrophils. While our knowledge of the biological effects of human NAP-l/IL-8 is sufficient to warrant four recent reviews5-8 and studies of clinical specimens suggest the involvement of NAPl/IL-8 in several human diseases such as psoriasis, rheumathoid arthritis, and adult respiratory distress syndrome,’ the elucidation of the in vivo pathophysiological role of this cytokine was hampered by the fact that animal molecules equivalent to human NAP-l/ IL-8 have not been yet identified. The unknown crossspecies potency of human NAP-l/IL-8 limited the interpretation and extrapolation to the human situation of the results obtained when human NAP-l/IL-8 was introduced into in vivo rabbit, rat, and mouse models.3~10-‘4 Here we establish the cross-species potency of rhNAP-l/IL-8 by comparing the in vitro chemotactic responses of isolated chicken, dog, goat, guinea-pig, monkey, mouse, pig, rabbit, and rat blood PMNLs. The chemotactic efficacy of NAP-l/IL-8 for PMNLs from all of the species was compared to that of zymosanactivated serum preparations (C5a) and formylmethionylisoleucylphenylalanylleucine (fMIFL). In vitro results were validated by the morphological evaluation of intradermal rhNAP-l/IL-8 injection sites in rabbits and rats.
21
22 / AntalRot
RESULTS ChemotaxisIn vitro The neutrophils* of all the tested species migrated to rhNAP-l/IL-g. Its optimal chemotactic concentration ranged from 37 ug/ml for rat neutrophils to 460 rig/ml for monkey neutrophils (Fig. 1 A, B, and C). Thus, the potency of rhNAP-l/IL-8 for animal neutrophils did not reach that for human cells (an optimal chemotactic concentration of 150 @ml). Table 1 shows the relative potency of rhNAP-l/IL-8 calculated on the basis of its ED,, for neutrophils of each species. The chemotactic efficacy of rhNAP-l/IL-8 for neutrophils from different species also differed, ranging from 15% to 80% (note the differing ordinate scales of the graphs A, B, and C in Fig. 1). These differences in the magnitude of chemotactic response to rhNAP-l/IL-8 could reflect the conditions of the assay not being optimal for the neutrophils of one or another species. For example, the use of a larger pore size filter or a longer incubation resulted in the same potency but higher chemotactic efficacy of rhNAP-l/IL-8 for the goat and pig neutrophils (Figs. 2, 3). The comparison of rhNAP-l/IL-8 efficacy to that of other known attractants (fMIFL, C5a) provided a parameter independent of the assay conditions. The ratios of rhNAPl/IL-8 and C5a efficacies were calculated because neutrophils of several species did not migrate to the formylated peptide (Table 2). The rhNAP-l/IL-8:CSa efficacy ratios for neutrophils from different species also varied. The efficacy ratio was close to 1 for the PMNLs of chicken, dog, guinea-pig, monkey, mouse, and human. The rhNAP-l/IL-8 attracted more than twice as many goat and pig neutrophils as C5a. Neutrophils from both of these species failed to migrate to the formylated peptide. The magnitude of the chemotactic response to rhNAP-l/IL-8 was less than that to C5a in the cases of rat and rabbit neutrophils (Table 2).
In viva E$ect of rhNAplIIL-8 The intradermal injections of rhNAP-l/IL-8 caused, in both rabbit and rat, dose-dependent neutrophil margination and massive infiltration, mostly around postcapillary venules of the lower dermis. The rabbit was more than ten times more sensitive than the rat (Fig. 4, A and B). This correlates well with the approximately 20-fold difference in the chemotactic
*For reasons of simplicity, we refer to the PMNLs of all the tested species as “neutrophils”, although from chicken blood we separated and tested heterophils (pseudoeosinophils), the avian equivalents of neutrophils. The term “heterophils” has also beenapplied to rabbit and guinea-pig PMNLs.r5
CYTOKJNE, Vol. 3, No. 1 (January 1991: 21-27)
potency of rhNAP-l/IL-8 for rabbit and rat (Table 1). In both species, the control injections of PBS resulted in the margination and extravasation of lymphocytes. In rats, lower rhNAP-l/IL-8 concentrations caused lymphocyte infiltration significantly exceeding that seen in the control sites.
DISCUSSION Granule-containing cells with morphology and function analogous to those of human neutrophils are present in a broad range of experimental species.16 The similarities between human and animal neutrophils substantiate the projection of the results from animal experimental studies on neutrophil pathophysiology into the human situation. However, profound interspeties differences exist in absolute and relative numbers of neutrophils and in all aspects of their function, including their responses to chemoattractants and cytokines.16 For example, formylated peptides, potent chemoattractants for human leukocytes, are not chemotactic for cow,17 pig,” dog,” cat,” and horse’l neutrophils. Here we have extended the list of animal PMNLs that do not respond to formylated peptides, showing that neither goat neutrophils nor chicken heterophils migrate to fMIFL. Another prototype chemoattractant for human neutrophils, LTB4, is not chemotactic for rat neutrophils.” Also, some human cytokines (e.g. IL-3, IL-4, granulocyte-macrophage colony-stimulating factor, etc.) are inactive in animals. Therefore, the introduction of NAP-l/IL-g, a human chemotactic cytokine, into an experimental animal model and the extrapolation of the findings to the human situation requires careful investigation of the interspecies potency of such a molecule. Besides neutrophils, NAP-l/ IL-8 attracts lymphocytes3x4 and basophils.4 Also, human monocytes and eosinophils have receptors for NAP-1/IL-8,23 but their function is not yet known. Comparison of the in vitro potency and efficacy of NAP-l/IL-8 for different human leukocytes showed that the primary proinflammatory action of this cytokine is on neutrophils.4 By measuring the in vitro chemotactic response of the neutrophils separated from the blood of different species we could ascertain the responsiveness of these species to the human cytokine. The identical PMNL separation procedure and conditions of the chemotaxis assay allowed direct comparison of the chemotactic potency of rhNAP-l/ IL-8 for the neutrophils of different species. The arbitrarily chosen parameters of the in vitro chemotaxis assay, including the filter pore size and the duration of the assay, influenced the chemotactic efficacy, but not potency, of rhNAP-l/IL-8 for neutrophils of different species. Thus, the observed potency
Species specificity of rhNAP-l/IL-8
/ 23
60 % 5 z +e E k la z .-en E VI 5 0
50
40
30
20
10
Ok 0,oi
LLLL
1
O,l rhNAP-l/IL-E -tf
Mouse
-e-
Rat
IO Concentration -A-
100
0 0,Ol
1000
Guinea-pig
+S- Goat
-e-
C
1.
Migration
of neutrophils
z .E =*
40
differences do not reflect unoptimized experimental conditions for neutrophils of one or another species but, rather, represent the differences in the ability of the human molecule to stimulate animal leukocytes via their receptors for the animal analogues of NAP-l/ IL-8. The difference between in vitro rhNAP-l/IL-8 potencies for rat and rabbit neutrophils was proportionate to that found in vitro.
+
Dog
1000
-A-
Rabbit
s
20 I
/
O,l
to rhNAP-l/%8.
(A) mouse, rat, guinea-pig, and goat; (B) monkey, dog, and rabbit; (C) chicken, human, and pig. The arrows on graphs A and B indicate the optimal chemotactic concentration of rhNAP-l/IL-8 for human neutrophils.
Monkey
100 @g/ml)
100,
B
Figure
1 10 RI rhNAP-l/IL-8 Concentration
(pg/ml)
I
--Ef
1
10
rhNAP-l/IL-8
Concentration
Chicken
-&
Human
100 (pg/ml)
++
Pig
In agreement with the previously published data,3.” the intradermal injections of high concentrations of rhNAF’-l/IL-8 resulted in neutrophil rich leukocyte infiltrates both in rabbit and rat, whereas low concentrations of rhNAP-l/IL-8 induced lymphocyte infiltration in rat but not rabbit. The observation that the injection of vehicle alone also causes massive lymphocyte emigration in both tested species deserves further
24 I Antal
CYTOKINE, Vol. 3, No. 1 (January1991:21-27)
Rot
Table 1. Relative potency of rhNAP-l/IL-f3 of different species.
for neutrophils
Species
Relativepotency
Human Monkey Guineapig Rabbit Dog Pig Goat Mouse Chicken Rat
100 34 27 19 11 8 7 4 4 1
investigation and is possibly due to the pressureinduced release of lymphocyte chemotaxins from resident cells, e.g., mast cells. Higher concentrations of rhNAP-l/IL-8 possibly have an inhibitory effect on lymphocyte emigration. In conclusion, in spite of the fact that rhNAP-l/ IL-8 is capable of attracting PMNLs from diffrent animal species, profound differences exist in its chemotactic potency for animal neutrophils. These differences should be accounted for when human NAP-l/ IL-8 is introduced into experimental animal models.
MATERIALS AND METHODS Experimental
Animals and Blood Sampling
100,
I
!
00,017 lo17
0,05
0,15
0,46
rhNAP-l/IL-8
+3ym
I,3
4.1
Concentration
12,3
1 i lLILLL 0.1
’ lIN’ff,
rhNAP-1IIL-8 +0
30 min. HBSS 30 min
1
1 ltt”’
1 10 Concentration(pglml) +
I I li I 100
45 min. 0
Figure 3. Migration of pig neutrophils different incubation periods.
HBSS45min
to rhNAP-l/IL-8
during
for the accommodation and care of animals by the Council of European Communities. Blood was drawn as shown in Table 3.
Separation of PMNLs
All experimental species were housed in the animal facility of the SF1and kept in accordance with the guidelines
80
0
37
111
The PMNLs from the heparinized blood of all the species studied and the venous blood of the normal human volunteers were separated by a slight modification of a differential centrifugation method.24 Briefly, 4 ml of blood were added to 16 ml of ice-cold 0.86 % ammonium cloride (Tris buffered, pH 7.4), mixed, and incubated for 15 min. Subsequently, 30 ml of Hanks’ balanced salt solution (BSS) were added to each tube and the cell suspensions were centrifuged at 18Ogfor 10 min. The supernatant was carefully discarded and the pellet was gently resuspended in 20 ml of Hanks’ BSS and centrifuged again at 16Ogfor 10 min. The resulting pellet was resuspended in 10 ml of Hanks’ BSS and centrifuged at 55g for 10 min. The cell pellet was resuspended in 5 ml of Hanks’ BSS and the final 5 min centrifugation was performed at 55g. The pellets from tubes processed in parallel were pooled; then absolute and differential counts were performed. The final preparations contained from more than 50% (guinea-pig, mouse, rat) to more than 80% (human, monkey, pig) viable neutrophils, as determined by a differential count of acridine orange-stained cells. For all species,the remaining cells in the preparation were lymphocytes. The neutrophil concentration was adjusted to 106/ml.
(yg/ml)
In FTtro Chemotaxis Assay
+-5ym
Figure 2. Migration of goat neutrophils to rhNAP-l/IL-8 the polycarbonate filters with different pore sizes.
through
The neutrophil chemotaxis assaywas performed using a 48-well modified Boyden chamber (Neuroprobe, Cabin John, MD, USA) and 3-urn pore size PVP-free polycarbonate
Species
specificity
of rhNAP-l/IL-8
/ 25
Table 2. Chemotactic efficacy* of rhNAP-l/IL-g, C5a, and fMIFL for neutrophils from different species. Chemotactic Species Chicken Dog Goat Guinea Monkey Mouse Pig Rabbit Rat Human
rhNAP-l/IL-8
*Efficacy:
C5a
fMIFL
55 58 7 24 81 27 35 81 47 67
2 2 1 45 46 26 1 94 25 72
58 60 15 24 60 30 80 36 17 63
pig
percentage
of cells migrating
at the optimal NAP-l/IL-8
efficacy (%) Hanks’ BSS
Efficacy ratio NAP-l/IL-8:CSa
2 1 0 2 I 7 1 8 2 5
1.1 1.0 2.1 1.0 0.8 1.1 2.3 0.4 0.4 0.9
concentration.
filters (Nucleopore, Pleasanton, CA, USA) as described before.” The following chemoattractants were used: NAP-I/IL-8. RhNAP-l/IL-8 was cloned, expressed, and purified as described before% and was generously provided by E. Wassebauer and I. Lindley (SFI, Vienna, Austria). Serial
three-fold dilutions of it were made in Hanks’ BSS containing Ca+’ and Mg+’ (Media Kitchen, SFI, Vienna, Austria). The highest tested concentration was 111 &ml. C5u. The zymosan-activated, heat-inactivated serum of each tested species (except monkey and human) was used
A
**
160
1q
LYMPHOCYTES
1
20 0
PBS
O,l
1
rhNAP-l/IL-8
6
70
10
100
concentration
1000
4.
Numbers
of leukocytes
in injection
sites.
(A) Number of leukocytes in the intradermal rhNAP-l/IL-8 injection sites in rabbit. (B) Number of leukocytes in the intradermal rhNAP-l/IL-8 injection sites in rat. Error bars show SEM for 10 field counts and asterisks indicate values significantly higher than the controls (*, p < 0.01; **, p < 0.001).
i
100000
**
1
60
Figure
10000
(nglml)
2o 10
0
PBS
or1
1
rhNAP-l/IL-8
concentration
(rig/ml)
26 / Antal Rot
CYTOKINE,
Vol. 3, No. 1 (January 1991: 21-27)
Table 3. Blood sampling for neutrophil separation. Species
Sex
Weight
Chicken Dog Goat Guinea pig Monkev Mouse’ Pig Rabbit Rat
M M F F F M M F F
12 kg 25 kg 250 g 9 kg 20-25 g 20 kg 4 kg 250 g
5kg
Strain
Blood source
Volume
Vedette Beagle SF1 breed Albino Rhesus BalbiC Landrace New Zealand White Wistar
Cardiac puncture Cephalic vein Jugular vein Cardiac puncture Brachial vein Cardiac puncture Jugular vein Central auricular artery Cardiac puncture
20 ml 20 ml 10 ml 8ml 10 ml 1 ml 20 ml 20 ml 8ml
threefold dilutions of activated serum preparations were used in chemotaxis assays,starting with a l/3 dilution. In human and monkey chemotaxis assays, human purified CSa (Sigma, Deisenhofen, Germany) was used in serial tenfold dilutions starting with 10 pg/ml. Formyluted Peptide. fMIPL (Peninsula Labs, Belmont, CA, USA), a formylated tertapeptide with high potency and efficacy,” was used in serial tenfold dilutions, starting with lo-’ M. Hanks’ BSS containing Ca+’ and Mg+’ was used as a negative control. Each attractant was tested in duplicate wells. The chambers were disassembled and filters removed after a 25-min incubation. On several occasions, the effects of longer assay time and filters with a larger pore size (5 km) were investigated. Neutrophils that migrated to the lower surface of the membrane were counted by an Optomax V image analyser (AiTektron, Meerbusch, Germany). Results were expressed as a percent of the input number of neutrophils that migrated per well for duplicate wells. The standard error of the mean (SEM) was less than 15%. as a source of C5a. Serial
In viva Eflect of rhNAp-l/IL-8 The leukocyte accumulation in the rabbit and rat skin injection sites was evaluated as follows. Serial tenfold dilutions of rhNAP-l/IL-8 were made in PBS (LPS < 1 pg/ml, Media Kitchen, SFI, Vienna, Austria). The highest concentration was 100 &ml and the lowest was 100 pg/ml. The fur was removed by clippers 24 h prior to injection, animals were anesthetized, and 100 ~1 of each rhNAP-l/IL-8 dilution and PBS were injected intradermally into duplicate sites using 26-gauge hypodermic needles. The center of each injection lump and the point 2 mm distal of each needle insertion were marked. Three hours following the injections, the rats and rabbits were sacrificed. The injection sites were cut out, attached to pieces of cardboard and fixed overnight in formalin. Following the fixation, injection sites were cut in half by a blade in the plane perpendicular to the skin surface and connecting the two markings. The resulting two halves were embedded in Paraplast (Monoject Scientific, Kildare, Ireland). Sections 3.5 pm thick were prepared on an automatic microtome (Ultracut 2050, Reichert-Jung, Vienna, Austria) and stained with alcian blue-H&E. The coded slides were studied under the microscope. The leukocytes in each section were counted in ten high-power fields (diameter 0.6 mm). The characteristic nuclear morphology and bright blue cytoplasm provided the basis for the identification of neutro-
phils and mast cells, respectively. Only the small round cells were counted as lymphocytes, whereas larger mononuclear cells could not be evaluated due to the heterogenous nature of this morphological entity. The results were expressed as a mean number of leukocytesper high-power field. The KruskalWallis statistical analysis (H-test) was performed, followed by multiple comparisons of NemCnyi. Acknowledgments The excellent technical assistance of Frau Marion
ZsAk and Herr Kamillo Thierer, the fruitful discussions with Drs. C. Lam and I. Lindley, and the help of Dr. F-P. Schmook in performing the statistical analysis are gratefully acknowledged.
REFERENCES 1. Galin JI, Wright DG, Malech HL, Davis JM, Klempner MS, Kirkpatrick CH (1980) Disorders of phagocyte chemotaxis. Ann Intern Med 92:520-538. 2. Yoshimura T, Matsushima K, Tanaka S, Robinson EA, Appella E, Oppenheim JJ, Leonard EJ (1987) Purification of a human monocyte-derived neutrophil chemotactic factor that shares sequence homology with other host defense cytokines. Proc Nat1 Acad Sci USA 84:9233-9237. 3. Larsen CG, Anderson AO, Appella E, Oppenheim JJ, Matsushima K (1989) The neutrophil-activating protein (NAP-l) is also chemotactic for T lymphocytes. Science 243:1464-1466. 4. Leonard EJ, Skeel A, Yoshimura T, Noer K, Kutvirt S, Van Epps D (1990) Leukocyte specificity and binding of human neutrophi1 attractant/activation protein-l (NAP-l). J Immunol 144:13231330. 5. Westwick J, Li SW, Camp RD (1989) Novel neutrophilstimulating peptides. Immunol Today 10:146-147. 6. Baggiolini M, Walz A, Kunkel SL (1989) Neutrophilactivating peptide-l/interleukin-8, a novel cytokine that activates neutrophils. J Clin Invest 84:1045-1049. 7. Matsushima K, Oppenheim JJ (1989) Interleukin 8 and MCAF: novel inflammatory cytokines inducible by IL-1 and TNF. Cytokine 1:2-13. 8. Leonard EJ, Yoshimura T (1990) Neutrophil attractant/ activation protein-l [NAP-l (IL-8)]. Am J Resp Cell Mol Biol 2~479-486. 9. Weshvick J, Kunkel S, Lindley IJD (eds) (1991) Chemotactic Cytokines: Biology of the Inflammatory Peptide Supergene Family. Plenum, London. 10. Van Damme J, Van Beeumen J, Opdenakker G, Billiau A (1988) A novel, NH,-terminal sequence-characterized human monokine possessing neutrophil chemotactic, skin reactive, and granulocytosis-promoting activity. J Exp Med 167:1364-1376.
Species specificity of rhNAP-l/IL-S 11. Colditz I, Zwahlen R, Dewald B, Baggiolini M (1989) In vivo inflammatory activity of neutrophil-activating factor, a novel chemotactic peptide derived from human monocytes. Am J Path01 134:755-760. 12. Rampart M, Van Damme J, Zonnekeyn L, Herman AG (1989) Granulocyte chemotactic protein/interleukin-8 induces plasma leakage and neutrophil accumulation in rabbit skin. Am J Pathol 135:21-26. 13. Furuta R, Yamagishi J, Kotani H, Sakamoto F, Fukui T, Matsui Y, Sohmura Y, Yamada M, Yoshimura T, Larsen CG, Oppenheim JJ, Matsushima K (1989) Production and characterisation of recombinant human neutrophil chemotactic factor. J Biothem 106:436-441. 14. Foster A, Aked DM, Schroder J-M, Christophers E (1989) Acute inflammatory effects of a monocyte-derived neutrophilactivating peptide in rabbit skin. Immunology 67:181-183. 15. Murata F, Spicer SS (1973) Morphologic and cytochemical studies of rabbit heterophilic leukocytes. Lab Invest 29:65-72. 16. Styrt B (1989) Species variation in neutrophil biochemistry and function. J Leukocyte Biol46:63-74. 17. Forsell JH, Kateley JR, Smith CW (1985) Bovine neutrophils treated with chemotactic agents: Morphologic changes. Am J Vet Res 46:1971-1985. 18. Chenoweth DE, Lane TA, Rowe JG, Hugli TE (1980) Quantitative comparisons of neutrophil chemotaxis in four animal species. Clin Immunol Immunopatho115:525-535. 19. Stickle JE, Kwan D, Smith CW (1985) Neutrophil function in the dog: Shape change and response to a synthetic tripeptide. Am J Vet Res 46:225-228. 20. Gray GD, Ohlmann GM, Morton DR, Schaub RG (1986) Feline polymorphonuclear leukocytes respond chemotactically to
/ 27
leukotriene B and activated serum but not to F-met-leu-phe. Agents Actions 18:401-407. 21. Snyderman R, Pike MC (1980) N-formylmethionyl peptide receptors on equine leukocytes initiate secretion but not chemotaxis. Science 209:493-495. 22. Kreisle RA, Parker CW, Griffin GL, Senior RM, Stenson WF (1985) Studies of leukotriene B4-specific binding and function in rat polymorphonuclear leukocytes: Absence of a chemotactic response. J Immunol134:3356-3363. 23. Besemer J, Hujber A, Kuhn B (1989). Specific binding, internalization, and degradation of human neutrophil activating factor by human polymorphonuclear leukocytes. J Biol Chem 264: 17409-17415. 24. Eggleton P, Gargan R, Fisher D (1989) Rapid method for the isolation of neutrophils in high yield without the use of dextran or density gradient polymers. J Immunol Methods 121:105-113. 25. Harvath L, Falk W, Leonard EJ (1980) Rapid quantitation of neutrophil chemotaxis: use of polyvinylpyrrolidone-free polycarbonate membrane in a multiwell assembly. J Immunol Methods 37:39-45. 26. Lindley I, Aschauer H, Seifert J-M, Lam C, Brunowsky W, Kownatsky E, Thelen M, Peveri P, Dewald B, von Tschamer V, Walz A, Baggiolini M (1988): Synthesis and expression in Escherichia coli of the gene encoding monocyte-derived neutrophil-activating factor: Biological equivalence between natural and recombinant neutrophil activating factor. Proc Nat1 Acad Sci USA 85:9199-9203. 27. Rot A, Henderson LE, Copeland TD, Leonard EJ (1987) A series of six ligands for the human formyl peptide receptor: Tetrapeptides with high chemotactic potency and efficacy. Proc Nat1 Acad Sci USA 84:7967-7971.
The polymorphonuclear neutrophil granulocyte (PMNL or neutrophil) is the predominant human blood leukocyte. The appearance and accumulation of neutrophils in the tissues is one of the morphological hallmarks of acute inflammation. Upon stimulation the circulating neutrophils marginate, adhere to the endothelial lining of the postcapillary venules, and, driven by the gradient of the chemotaxin, migrate across the vessel wall into the tissues where at the site of inflammation they are triggered by high concentrations of inflammatory mediators to generate reactive oxygen radicals and secrete the contents of lysosomal granules.’ Whereas most of the known leukocyte chemotaxins attract both neutrophils and monocytes, the recently described neutrophil attractant/activation protein-l (interleukin-8, NAP-l/IL+ attracts neutrophils, lymphocytes, and basophils, but not monocytes.2-4 Thus, in contrast to the formylated peptides, C5a and LTB,, NAP-l/IL-8 could be the mediator of the inflammatory
Sandoz Forschungsinstitut, Copyright 0 1991 by W.B. 1043-4666/91/0301-0008$05.00/0 KEY WORDS: specificity
CYTOKINE,
Brunner Saunders
Str. 59, Vienna Company
chemotaxislinterleukin-8/IL-g/neutrophils/species
Vol. 3, No. 1 (January),
1991: pp 21-27
A-1235,
Austria.
responses dominated by neutrophils. While our knowledge of the biological effects of human NAP-l/IL-8 is sufficient to warrant four recent reviews5-8 and studies of clinical specimens suggest the involvement of NAPl/IL-8 in several human diseases such as psoriasis, rheumathoid arthritis, and adult respiratory distress syndrome,’ the elucidation of the in vivo pathophysiological role of this cytokine was hampered by the fact that animal molecules equivalent to human NAP-l/ IL-8 have not been yet identified. The unknown crossspecies potency of human NAP-l/IL-8 limited the interpretation and extrapolation to the human situation of the results obtained when human NAP-l/IL-8 was introduced into in vivo rabbit, rat, and mouse models.3~10-‘4 Here we establish the cross-species potency of rhNAP-l/IL-8 by comparing the in vitro chemotactic responses of isolated chicken, dog, goat, guinea-pig, monkey, mouse, pig, rabbit, and rat blood PMNLs. The chemotactic efficacy of NAP-l/IL-8 for PMNLs from all of the species was compared to that of zymosanactivated serum preparations (C5a) and formylmethionylisoleucylphenylalanylleucine (fMIFL). In vitro results were validated by the morphological evaluation of intradermal rhNAP-l/IL-8 injection sites in rabbits and rats.
21
22 / AntalRot
RESULTS ChemotaxisIn vitro The neutrophils* of all the tested species migrated to rhNAP-l/IL-g. Its optimal chemotactic concentration ranged from 37 ug/ml for rat neutrophils to 460 rig/ml for monkey neutrophils (Fig. 1 A, B, and C). Thus, the potency of rhNAP-l/IL-8 for animal neutrophils did not reach that for human cells (an optimal chemotactic concentration of 150 @ml). Table 1 shows the relative potency of rhNAP-l/IL-8 calculated on the basis of its ED,, for neutrophils of each species. The chemotactic efficacy of rhNAP-l/IL-8 for neutrophils from different species also differed, ranging from 15% to 80% (note the differing ordinate scales of the graphs A, B, and C in Fig. 1). These differences in the magnitude of chemotactic response to rhNAP-l/IL-8 could reflect the conditions of the assay not being optimal for the neutrophils of one or another species. For example, the use of a larger pore size filter or a longer incubation resulted in the same potency but higher chemotactic efficacy of rhNAP-l/IL-8 for the goat and pig neutrophils (Figs. 2, 3). The comparison of rhNAP-l/IL-8 efficacy to that of other known attractants (fMIFL, C5a) provided a parameter independent of the assay conditions. The ratios of rhNAPl/IL-8 and C5a efficacies were calculated because neutrophils of several species did not migrate to the formylated peptide (Table 2). The rhNAP-l/IL-8:CSa efficacy ratios for neutrophils from different species also varied. The efficacy ratio was close to 1 for the PMNLs of chicken, dog, guinea-pig, monkey, mouse, and human. The rhNAP-l/IL-8 attracted more than twice as many goat and pig neutrophils as C5a. Neutrophils from both of these species failed to migrate to the formylated peptide. The magnitude of the chemotactic response to rhNAP-l/IL-8 was less than that to C5a in the cases of rat and rabbit neutrophils (Table 2).
In viva E$ect of rhNAplIIL-8 The intradermal injections of rhNAP-l/IL-8 caused, in both rabbit and rat, dose-dependent neutrophil margination and massive infiltration, mostly around postcapillary venules of the lower dermis. The rabbit was more than ten times more sensitive than the rat (Fig. 4, A and B). This correlates well with the approximately 20-fold difference in the chemotactic
*For reasons of simplicity, we refer to the PMNLs of all the tested species as “neutrophils”, although from chicken blood we separated and tested heterophils (pseudoeosinophils), the avian equivalents of neutrophils. The term “heterophils” has also beenapplied to rabbit and guinea-pig PMNLs.r5
CYTOKJNE, Vol. 3, No. 1 (January 1991: 21-27)
potency of rhNAP-l/IL-8 for rabbit and rat (Table 1). In both species, the control injections of PBS resulted in the margination and extravasation of lymphocytes. In rats, lower rhNAP-l/IL-8 concentrations caused lymphocyte infiltration significantly exceeding that seen in the control sites.
DISCUSSION Granule-containing cells with morphology and function analogous to those of human neutrophils are present in a broad range of experimental species.16 The similarities between human and animal neutrophils substantiate the projection of the results from animal experimental studies on neutrophil pathophysiology into the human situation. However, profound interspeties differences exist in absolute and relative numbers of neutrophils and in all aspects of their function, including their responses to chemoattractants and cytokines.16 For example, formylated peptides, potent chemoattractants for human leukocytes, are not chemotactic for cow,17 pig,” dog,” cat,” and horse’l neutrophils. Here we have extended the list of animal PMNLs that do not respond to formylated peptides, showing that neither goat neutrophils nor chicken heterophils migrate to fMIFL. Another prototype chemoattractant for human neutrophils, LTB4, is not chemotactic for rat neutrophils.” Also, some human cytokines (e.g. IL-3, IL-4, granulocyte-macrophage colony-stimulating factor, etc.) are inactive in animals. Therefore, the introduction of NAP-l/IL-g, a human chemotactic cytokine, into an experimental animal model and the extrapolation of the findings to the human situation requires careful investigation of the interspecies potency of such a molecule. Besides neutrophils, NAP-l/ IL-8 attracts lymphocytes3x4 and basophils.4 Also, human monocytes and eosinophils have receptors for NAP-1/IL-8,23 but their function is not yet known. Comparison of the in vitro potency and efficacy of NAP-l/IL-8 for different human leukocytes showed that the primary proinflammatory action of this cytokine is on neutrophils.4 By measuring the in vitro chemotactic response of the neutrophils separated from the blood of different species we could ascertain the responsiveness of these species to the human cytokine. The identical PMNL separation procedure and conditions of the chemotaxis assay allowed direct comparison of the chemotactic potency of rhNAP-l/ IL-8 for the neutrophils of different species. The arbitrarily chosen parameters of the in vitro chemotaxis assay, including the filter pore size and the duration of the assay, influenced the chemotactic efficacy, but not potency, of rhNAP-l/IL-8 for neutrophils of different species. Thus, the observed potency
Species specificity of rhNAP-l/IL-8
/ 23
60 % 5 z +e E k la z .-en E VI 5 0
50
40
30
20
10
Ok 0,oi
LLLL
1
O,l rhNAP-l/IL-E -tf
Mouse
-e-
Rat
IO Concentration -A-
100
0 0,Ol
1000
Guinea-pig
+S- Goat
-e-
C
1.
Migration
of neutrophils
z .E =*
40
differences do not reflect unoptimized experimental conditions for neutrophils of one or another species but, rather, represent the differences in the ability of the human molecule to stimulate animal leukocytes via their receptors for the animal analogues of NAP-l/ IL-8. The difference between in vitro rhNAP-l/IL-8 potencies for rat and rabbit neutrophils was proportionate to that found in vitro.
+
Dog
1000
-A-
Rabbit
s
20 I
/
O,l
to rhNAP-l/%8.
(A) mouse, rat, guinea-pig, and goat; (B) monkey, dog, and rabbit; (C) chicken, human, and pig. The arrows on graphs A and B indicate the optimal chemotactic concentration of rhNAP-l/IL-8 for human neutrophils.
Monkey
100 @g/ml)
100,
B
Figure
1 10 RI rhNAP-l/IL-8 Concentration
(pg/ml)
I
--Ef
1
10
rhNAP-l/IL-8
Concentration
Chicken
-&
Human
100 (pg/ml)
++
Pig
In agreement with the previously published data,3.” the intradermal injections of high concentrations of rhNAF’-l/IL-8 resulted in neutrophil rich leukocyte infiltrates both in rabbit and rat, whereas low concentrations of rhNAP-l/IL-8 induced lymphocyte infiltration in rat but not rabbit. The observation that the injection of vehicle alone also causes massive lymphocyte emigration in both tested species deserves further
24 I Antal
CYTOKINE, Vol. 3, No. 1 (January1991:21-27)
Rot
Table 1. Relative potency of rhNAP-l/IL-f3 of different species.
for neutrophils
Species
Relativepotency
Human Monkey Guineapig Rabbit Dog Pig Goat Mouse Chicken Rat
100 34 27 19 11 8 7 4 4 1
investigation and is possibly due to the pressureinduced release of lymphocyte chemotaxins from resident cells, e.g., mast cells. Higher concentrations of rhNAP-l/IL-8 possibly have an inhibitory effect on lymphocyte emigration. In conclusion, in spite of the fact that rhNAP-l/ IL-8 is capable of attracting PMNLs from diffrent animal species, profound differences exist in its chemotactic potency for animal neutrophils. These differences should be accounted for when human NAP-l/ IL-8 is introduced into experimental animal models.
MATERIALS AND METHODS Experimental
Animals and Blood Sampling
100,
I
!
00,017 lo17
0,05
0,15
0,46
rhNAP-l/IL-8
+3ym
I,3
4.1
Concentration
12,3
1 i lLILLL 0.1
’ lIN’ff,
rhNAP-1IIL-8 +0
30 min. HBSS 30 min
1
1 ltt”’
1 10 Concentration(pglml) +
I I li I 100
45 min. 0
Figure 3. Migration of pig neutrophils different incubation periods.
HBSS45min
to rhNAP-l/IL-8
during
for the accommodation and care of animals by the Council of European Communities. Blood was drawn as shown in Table 3.
Separation of PMNLs
All experimental species were housed in the animal facility of the SF1and kept in accordance with the guidelines
80
0
37
111
The PMNLs from the heparinized blood of all the species studied and the venous blood of the normal human volunteers were separated by a slight modification of a differential centrifugation method.24 Briefly, 4 ml of blood were added to 16 ml of ice-cold 0.86 % ammonium cloride (Tris buffered, pH 7.4), mixed, and incubated for 15 min. Subsequently, 30 ml of Hanks’ balanced salt solution (BSS) were added to each tube and the cell suspensions were centrifuged at 18Ogfor 10 min. The supernatant was carefully discarded and the pellet was gently resuspended in 20 ml of Hanks’ BSS and centrifuged again at 16Ogfor 10 min. The resulting pellet was resuspended in 10 ml of Hanks’ BSS and centrifuged at 55g for 10 min. The cell pellet was resuspended in 5 ml of Hanks’ BSS and the final 5 min centrifugation was performed at 55g. The pellets from tubes processed in parallel were pooled; then absolute and differential counts were performed. The final preparations contained from more than 50% (guinea-pig, mouse, rat) to more than 80% (human, monkey, pig) viable neutrophils, as determined by a differential count of acridine orange-stained cells. For all species,the remaining cells in the preparation were lymphocytes. The neutrophil concentration was adjusted to 106/ml.
(yg/ml)
In FTtro Chemotaxis Assay
+-5ym
Figure 2. Migration of goat neutrophils to rhNAP-l/IL-8 the polycarbonate filters with different pore sizes.
through
The neutrophil chemotaxis assaywas performed using a 48-well modified Boyden chamber (Neuroprobe, Cabin John, MD, USA) and 3-urn pore size PVP-free polycarbonate
Species
specificity
of rhNAP-l/IL-8
/ 25
Table 2. Chemotactic efficacy* of rhNAP-l/IL-g, C5a, and fMIFL for neutrophils from different species. Chemotactic Species Chicken Dog Goat Guinea Monkey Mouse Pig Rabbit Rat Human
rhNAP-l/IL-8
*Efficacy:
C5a
fMIFL
55 58 7 24 81 27 35 81 47 67
2 2 1 45 46 26 1 94 25 72
58 60 15 24 60 30 80 36 17 63
pig
percentage
of cells migrating
at the optimal NAP-l/IL-8
efficacy (%) Hanks’ BSS
Efficacy ratio NAP-l/IL-8:CSa
2 1 0 2 I 7 1 8 2 5
1.1 1.0 2.1 1.0 0.8 1.1 2.3 0.4 0.4 0.9
concentration.
filters (Nucleopore, Pleasanton, CA, USA) as described before.” The following chemoattractants were used: NAP-I/IL-8. RhNAP-l/IL-8 was cloned, expressed, and purified as described before% and was generously provided by E. Wassebauer and I. Lindley (SFI, Vienna, Austria). Serial
three-fold dilutions of it were made in Hanks’ BSS containing Ca+’ and Mg+’ (Media Kitchen, SFI, Vienna, Austria). The highest tested concentration was 111 &ml. C5u. The zymosan-activated, heat-inactivated serum of each tested species (except monkey and human) was used
A
**
160
1q
LYMPHOCYTES
1
20 0
PBS
O,l
1
rhNAP-l/IL-8
6
70
10
100
concentration
1000
4.
Numbers
of leukocytes
in injection
sites.
(A) Number of leukocytes in the intradermal rhNAP-l/IL-8 injection sites in rabbit. (B) Number of leukocytes in the intradermal rhNAP-l/IL-8 injection sites in rat. Error bars show SEM for 10 field counts and asterisks indicate values significantly higher than the controls (*, p < 0.01; **, p < 0.001).
i
100000
**
1
60
Figure
10000
(nglml)
2o 10
0
PBS
or1
1
rhNAP-l/IL-8
concentration
(rig/ml)
26 / Antal Rot
CYTOKINE,
Vol. 3, No. 1 (January 1991: 21-27)
Table 3. Blood sampling for neutrophil separation. Species
Sex
Weight
Chicken Dog Goat Guinea pig Monkev Mouse’ Pig Rabbit Rat
M M F F F M M F F
12 kg 25 kg 250 g 9 kg 20-25 g 20 kg 4 kg 250 g
5kg
Strain
Blood source
Volume
Vedette Beagle SF1 breed Albino Rhesus BalbiC Landrace New Zealand White Wistar
Cardiac puncture Cephalic vein Jugular vein Cardiac puncture Brachial vein Cardiac puncture Jugular vein Central auricular artery Cardiac puncture
20 ml 20 ml 10 ml 8ml 10 ml 1 ml 20 ml 20 ml 8ml
threefold dilutions of activated serum preparations were used in chemotaxis assays,starting with a l/3 dilution. In human and monkey chemotaxis assays, human purified CSa (Sigma, Deisenhofen, Germany) was used in serial tenfold dilutions starting with 10 pg/ml. Formyluted Peptide. fMIPL (Peninsula Labs, Belmont, CA, USA), a formylated tertapeptide with high potency and efficacy,” was used in serial tenfold dilutions, starting with lo-’ M. Hanks’ BSS containing Ca+’ and Mg+’ was used as a negative control. Each attractant was tested in duplicate wells. The chambers were disassembled and filters removed after a 25-min incubation. On several occasions, the effects of longer assay time and filters with a larger pore size (5 km) were investigated. Neutrophils that migrated to the lower surface of the membrane were counted by an Optomax V image analyser (AiTektron, Meerbusch, Germany). Results were expressed as a percent of the input number of neutrophils that migrated per well for duplicate wells. The standard error of the mean (SEM) was less than 15%. as a source of C5a. Serial
In viva Eflect of rhNAp-l/IL-8 The leukocyte accumulation in the rabbit and rat skin injection sites was evaluated as follows. Serial tenfold dilutions of rhNAP-l/IL-8 were made in PBS (LPS < 1 pg/ml, Media Kitchen, SFI, Vienna, Austria). The highest concentration was 100 &ml and the lowest was 100 pg/ml. The fur was removed by clippers 24 h prior to injection, animals were anesthetized, and 100 ~1 of each rhNAP-l/IL-8 dilution and PBS were injected intradermally into duplicate sites using 26-gauge hypodermic needles. The center of each injection lump and the point 2 mm distal of each needle insertion were marked. Three hours following the injections, the rats and rabbits were sacrificed. The injection sites were cut out, attached to pieces of cardboard and fixed overnight in formalin. Following the fixation, injection sites were cut in half by a blade in the plane perpendicular to the skin surface and connecting the two markings. The resulting two halves were embedded in Paraplast (Monoject Scientific, Kildare, Ireland). Sections 3.5 pm thick were prepared on an automatic microtome (Ultracut 2050, Reichert-Jung, Vienna, Austria) and stained with alcian blue-H&E. The coded slides were studied under the microscope. The leukocytes in each section were counted in ten high-power fields (diameter 0.6 mm). The characteristic nuclear morphology and bright blue cytoplasm provided the basis for the identification of neutro-
phils and mast cells, respectively. Only the small round cells were counted as lymphocytes, whereas larger mononuclear cells could not be evaluated due to the heterogenous nature of this morphological entity. The results were expressed as a mean number of leukocytesper high-power field. The KruskalWallis statistical analysis (H-test) was performed, followed by multiple comparisons of NemCnyi. Acknowledgments The excellent technical assistance of Frau Marion
ZsAk and Herr Kamillo Thierer, the fruitful discussions with Drs. C. Lam and I. Lindley, and the help of Dr. F-P. Schmook in performing the statistical analysis are gratefully acknowledged.
REFERENCES 1. Galin JI, Wright DG, Malech HL, Davis JM, Klempner MS, Kirkpatrick CH (1980) Disorders of phagocyte chemotaxis. Ann Intern Med 92:520-538. 2. Yoshimura T, Matsushima K, Tanaka S, Robinson EA, Appella E, Oppenheim JJ, Leonard EJ (1987) Purification of a human monocyte-derived neutrophil chemotactic factor that shares sequence homology with other host defense cytokines. Proc Nat1 Acad Sci USA 84:9233-9237. 3. Larsen CG, Anderson AO, Appella E, Oppenheim JJ, Matsushima K (1989) The neutrophil-activating protein (NAP-l) is also chemotactic for T lymphocytes. Science 243:1464-1466. 4. Leonard EJ, Skeel A, Yoshimura T, Noer K, Kutvirt S, Van Epps D (1990) Leukocyte specificity and binding of human neutrophi1 attractant/activation protein-l (NAP-l). J Immunol 144:13231330. 5. Westwick J, Li SW, Camp RD (1989) Novel neutrophilstimulating peptides. Immunol Today 10:146-147. 6. Baggiolini M, Walz A, Kunkel SL (1989) Neutrophilactivating peptide-l/interleukin-8, a novel cytokine that activates neutrophils. J Clin Invest 84:1045-1049. 7. Matsushima K, Oppenheim JJ (1989) Interleukin 8 and MCAF: novel inflammatory cytokines inducible by IL-1 and TNF. Cytokine 1:2-13. 8. Leonard EJ, Yoshimura T (1990) Neutrophil attractant/ activation protein-l [NAP-l (IL-8)]. Am J Resp Cell Mol Biol 2~479-486. 9. Weshvick J, Kunkel S, Lindley IJD (eds) (1991) Chemotactic Cytokines: Biology of the Inflammatory Peptide Supergene Family. Plenum, London. 10. Van Damme J, Van Beeumen J, Opdenakker G, Billiau A (1988) A novel, NH,-terminal sequence-characterized human monokine possessing neutrophil chemotactic, skin reactive, and granulocytosis-promoting activity. J Exp Med 167:1364-1376.
Species specificity of rhNAP-l/IL-S 11. Colditz I, Zwahlen R, Dewald B, Baggiolini M (1989) In vivo inflammatory activity of neutrophil-activating factor, a novel chemotactic peptide derived from human monocytes. Am J Path01 134:755-760. 12. Rampart M, Van Damme J, Zonnekeyn L, Herman AG (1989) Granulocyte chemotactic protein/interleukin-8 induces plasma leakage and neutrophil accumulation in rabbit skin. Am J Pathol 135:21-26. 13. Furuta R, Yamagishi J, Kotani H, Sakamoto F, Fukui T, Matsui Y, Sohmura Y, Yamada M, Yoshimura T, Larsen CG, Oppenheim JJ, Matsushima K (1989) Production and characterisation of recombinant human neutrophil chemotactic factor. J Biothem 106:436-441. 14. Foster A, Aked DM, Schroder J-M, Christophers E (1989) Acute inflammatory effects of a monocyte-derived neutrophilactivating peptide in rabbit skin. Immunology 67:181-183. 15. Murata F, Spicer SS (1973) Morphologic and cytochemical studies of rabbit heterophilic leukocytes. Lab Invest 29:65-72. 16. Styrt B (1989) Species variation in neutrophil biochemistry and function. J Leukocyte Biol46:63-74. 17. Forsell JH, Kateley JR, Smith CW (1985) Bovine neutrophils treated with chemotactic agents: Morphologic changes. Am J Vet Res 46:1971-1985. 18. Chenoweth DE, Lane TA, Rowe JG, Hugli TE (1980) Quantitative comparisons of neutrophil chemotaxis in four animal species. Clin Immunol Immunopatho115:525-535. 19. Stickle JE, Kwan D, Smith CW (1985) Neutrophil function in the dog: Shape change and response to a synthetic tripeptide. Am J Vet Res 46:225-228. 20. Gray GD, Ohlmann GM, Morton DR, Schaub RG (1986) Feline polymorphonuclear leukocytes respond chemotactically to
/ 27
leukotriene B and activated serum but not to F-met-leu-phe. Agents Actions 18:401-407. 21. Snyderman R, Pike MC (1980) N-formylmethionyl peptide receptors on equine leukocytes initiate secretion but not chemotaxis. Science 209:493-495. 22. Kreisle RA, Parker CW, Griffin GL, Senior RM, Stenson WF (1985) Studies of leukotriene B4-specific binding and function in rat polymorphonuclear leukocytes: Absence of a chemotactic response. J Immunol134:3356-3363. 23. Besemer J, Hujber A, Kuhn B (1989). Specific binding, internalization, and degradation of human neutrophil activating factor by human polymorphonuclear leukocytes. J Biol Chem 264: 17409-17415. 24. Eggleton P, Gargan R, Fisher D (1989) Rapid method for the isolation of neutrophils in high yield without the use of dextran or density gradient polymers. J Immunol Methods 121:105-113. 25. Harvath L, Falk W, Leonard EJ (1980) Rapid quantitation of neutrophil chemotaxis: use of polyvinylpyrrolidone-free polycarbonate membrane in a multiwell assembly. J Immunol Methods 37:39-45. 26. Lindley I, Aschauer H, Seifert J-M, Lam C, Brunowsky W, Kownatsky E, Thelen M, Peveri P, Dewald B, von Tschamer V, Walz A, Baggiolini M (1988): Synthesis and expression in Escherichia coli of the gene encoding monocyte-derived neutrophil-activating factor: Biological equivalence between natural and recombinant neutrophil activating factor. Proc Nat1 Acad Sci USA 85:9199-9203. 27. Rot A, Henderson LE, Copeland TD, Leonard EJ (1987) A series of six ligands for the human formyl peptide receptor: Tetrapeptides with high chemotactic potency and efficacy. Proc Nat1 Acad Sci USA 84:7967-7971.
